Metrology Variability and its Impact in Process Modeling

In electron proximity effects correction (PEC), the quality of a correction is highly dependent on the quality of the model used to compute the effects. Therefore it is of primary importance to have a reliable methodology to extract the parameters and assess the quality of a model. Usually, model calibration procedures consist of one or more cycles of exposure and measurements on the calibration stage. The process and metrology variability may play a key role in the quality of the final model and, hence, of the PEC result. Therefore, it is important to determine at which level these variations may impact a calibration procedure and how a calibration design may be implemented in order to enable more robustness to the resulting model. In this work, metrology variability was evaluated by measuring the same wafer using two different CD-SEM tools. The information coming from these analyses was used as reference to a variation induced calibration test using synthetic data. By inserting variability in synthetic data it was possible to evaluate its impact on the resulting parameter values and in the final model error evaluation.